Hybrid recommender system using association rules
| aut.embargo | No | en |
| aut.thirdpc.contains | No | |
| aut.thirdpc.permission | No | |
| aut.thirdpc.removed | No | |
| dc.contributor.advisor | Pears, Russel | |
| dc.contributor.author | Cristache, Alex | |
| dc.date.accessioned | 2010-03-11T02:58:45Z | |
| dc.date.available | 2010-03-11T02:58:45Z | |
| dc.date.copyright | 2009 | |
| dc.date.issued | 2009 | |
| dc.date.updated | 2010-03-10T04:16:17Z | |
| dc.description.abstract | Recommender systems are increasingly being used in today’s world. Collaborative filtering, together with association rules mining are probably the most widely used methods to implement recommender systems. In this dissertation we undertake a review of past research conducted in the area of recommender systems with the focus being the use of association rule mining. We propose a novel methodology that combines the use of association mining with the use of distance metrics such as the Jaccard measure to identify movies that belong to the same genre. Our experimental results on the MovieLens dataset shows that the use of the Jaccard metric improved the coverage of recommendations over the use of the standard association rule mining method. | |
| dc.identifier.uri | https://hdl.handle.net/10292/822 | |
| dc.language.iso | en | en |
| dc.publisher | Auckland University of Technology | |
| dc.rights.accessrights | OpenAccess | |
| dc.subject | Recommender system | |
| dc.subject | Association rules | |
| dc.subject | Jaccard | |
| dc.subject | MovieLens | |
| dc.subject | Constructive research | |
| dc.subject | Computer science | |
| dc.title | Hybrid recommender system using association rules | |
| dc.type | Thesis | |
| thesis.degree.grantor | Auckland University of Technology | |
| thesis.degree.level | Masters Dissertations | |
| thesis.degree.name | Master of Computer and Information Sciences |